Understanding the cascade of genetic changes that leads to the formation of cancer promises a revolution in the prevention, early detection, and treatment of cancer by approaches that are innovative and specific. In addition, molecular profiling of tumors is likely to have far greater prognostic ability than traditional staging and grading. Advances in understanding early molecular genetic changes for ovarian cancer have been slow likely due to: a relative paucity of early-staged cancers for study of this disease, infrequent availability of fresh tissue for early-stage lesions, and histologic diversity of the disease which may confound efforts to identify common pathways. For this study, we will utilize innovative high throughput technologies for DNA and RNA analyses of early stage ovarian cancers obtained from two sources: 272 well-annotated formalin fixed specimens from the Gynecologic Oncology Group protocol 157 and 100 specimens collected from Partners Hospital with snap frozen tissue available. These specimens will be evaluated using the NCI eDNA array platform for comparative genomic hybridization analysis to detect DNA copy number abnormalities and by single nucleotide polymorphisms (SNP)-array-based loss of heterozygosity analysis on the 1OK SNP array platform to detect allelic loss profiles. Candidate genes will be first selected by performing expression profiling with the eDNA array platform and then further validated by fluorescent in situ hybridization (FISH), quantitative PCR, immunohistochemistry and enzyme linked immunosorbant assay (ELISA). All analyses will be histology-specific and use bioinformatics techniques suitable for assay based data to identify genetic changes correlated with clinical outcomes. Our hypothesis is that the identification of histologyspecific molecular genetic changes will suggest potential targets for prevention, early detection, or treatment strategies, and that the molecular genetic profiles will provide important predictors.